A heavy member can meet its specified compressive strength and, years later, open a map crack. The cylinder never saw what happened in the core during the first week: too much heat, at the wrong moment of hydration. The mechanism is delayed ettringite formation, DEF.
Early ettringite is not the problem
Ettringite is a normal product of Portland cement. In the first hours the calcium sulfate added to the cement reacts with the aluminate and forms ettringite, hydrated calcium trisulfoaluminate. That early ettringite regulates set. Because it forms while the paste is still plastic, it does not produce expansion damage. Mario Collepardi separates that early formation, cleanly, from the one that happens late, in paste that has already hardened.
The problem is not “having ettringite.” The problem is forming it again when the concrete has nowhere left to grow.
What the heat undoes
If the concrete passes about 70 °C at early age — from heat curing, or because the heat of hydration stayed trapped in the core — the early ettringite is no longer stable. Part of the sulfate and the aluminate is held in other phases, above all in the C-S-H, instead of forming ettringite at the time.
Years later, with the member cool and in contact with moisture, those ions leave where they were stored and ettringite recrystallizes inside the rigid paste. The review by Taylor, Famy and Scrivener describes the expansion as the effect of submicrometre crystals in the paste. The large crystals, easy to see in a crack or a void, are largely a recrystallization product, not the first cause of the swelling.
It is not a one-degree switch. The risk depends on the cement (aluminate, sulfate, alkalis), on how long the temperature stayed high, and on the moisture available afterwards. The line near 70 °C is the level the literature uses as a warning, not as an automatic sentence. Below it the risk falls. Above it, ignoring the peak is buying a long-term problem.
Why mass concrete is in this account
Mass concrete is any volume in which the heat of hydration does not leave as fast as it is generated. A foundation block, a thick footing, a transfer girder and a robust wall fall under that definition even when the drawings never used the name.
In a hot climate the path is shorter. The fresh concrete already arrives warmer, and the air does little to pull heat out of the core. Cuiabá does not need a record temperature for the care to make sense: a large member, too much cement and a hot placement are a known trio.
The peak that matters is the core, not the air and not the cylinder cast beside the member, which cools easily. Measuring only the air temperature measures the weather, not the member.
What the crack does not say by itself
DEF is not external sulfate attack, in which the aggressive ion comes from outside. It is not alkali–aggregate reaction either, although the crack can look the same and the mechanisms can coexist. Closing a diagnosis from a photograph of the crack is a guess.
What helps: the thermal history of the member, the cement that went in, whether there was heat curing, petrography and, where it fits, an expansion test. A compressive strength that was met does not clear the member. A 100 mm or 150 mm cylinder never lived the heat in the core of a one-metre block.
What lowers the risk
Treat the thick member as mass concrete from the placement plan. Estimate the peak and measure temperature inside the member.
Limit the heat released. A conscious cement content, supplementary cementitious materials where performance allows — fly ash, slag, filler, each in its own case — and an early strength that is not bought only with more cement.
Agree the suitable cement with the supplier when there is mass or heat curing. The balance between sulfate and aluminate stops being a factory detail and becomes a mix-design variable.
Dissipate on purpose: lifts, time between placements, shade, cooler mixing water, ice when the volume justifies it. Each measure belongs in the plan. On the day of the pour, a thermal improvisation is late.
DEF is an uncomfortable reminder. Concrete can pass the 28-day compressive strength test and still carry a chemical defect planted in the first week. A quality-control file that only archives that test did not see the heat.
References
- Taylor, H. F. W.; Famy, C.; Scrivener, K. L. Delayed ettringite formation. Cement and Concrete Research, v. 31, n. 5, p. 683–693, 2001. DOI: 10.1016/S0008-8846(01)00466-5.
- Collepardi, M. A state-of-the-art review on delayed ettringite attack on concrete. Cement and Concrete Composites, v. 25, n. 4–5, p. 401–407, 2003. DOI: 10.1016/S0958-9465(02)00080-X.
- Heinz, D.; Ludwig, U. Mechanism of secondary ettringite formation in mortars and concretes subjected to heat treatment. In: Concrete Durability: Katharine and Bryant Mather International Conference. ACI SP-100, 1987.
- American Concrete Institute. ACI 207.1R — Guide to Mass Concrete. Use the current edition.
- ABNT. NBR 12655 — Portland cement concrete — Preparation, control, receiving and acceptance. Use the current edition. The standard organizes production and control; it does not replace a thermal criterion for the member. Parts of ACI 301 and EN 206 also cover production and acceptance. The criteria are not interchangeable.